摘要
随着工控网络的发展以及工业和和信息化的深度融合,针对工业控制系统的攻击行为大幅度增长,对工控企业造成巨大的经济及财产损失。因此,提出一种基于半监督机器学习的入侵检测技术,该技术充分利用工控网络流量标记的特点,结合多种机器学习算法进行实现,并对算法的性能进行了优化。实验证明,该技术可以有效地检测出工控系统网络中的异常流量。
With the development of industry control system （ICS） network, especially the deep combination of industry and information tech- nology, attacks on ICS are greatly increased, which cause huge economic and property damage to ICS enterprises. In this paper, we design an intrusion detection method based on semi-supervised learning, which makes full use of the labelled feather of ICS network traffic, combines var- ied machine learning algorithms and improves the performance of the algorithms. Experimental results show that the method can effectively de- tect abnormal traffic of ICS network.
出处
《信息技术与网络安全》
2018年第1期44-47,共4页
Information Technology and Network Security
关键词
工业控制系统
入侵检测
半监督分类
机器学习
industry control system
intrusion detection
semi-supervised classification
machine learning
